Few events are as important to global understanding of human-induced environmental change as the American Dust Bowl. This application requests support to explore the relationship between environment and population loss in the American grasslands by developing a rich integration of information on human- environment systems. It extends a focus on measuring individual and household level change in new settlement areas, to the analysis of mature communities under conditions of climate stress and globalization. By implication, this approach demands a greater realism in understanding the economic factors and a long run view of the dynamics of change. To study the vulnerability of people and landscapes under stress, the project will pilot the creation a one-of-a-kind dataset on people, farms and landscapes: gathering individual level information from Kansas State agricultural census rolls, land parcel and mortgage lending information from deed registers, land cover data from a time series of aerial photographs, soil quality from digital soil surveys, and weather data from historical climate datasets. As drought revisited the region in the 1950s and 1970s, the State of Kansas continued to conduct annual censuses of agriculture. The proposed dataset will be the first of its kind to develop high-resolution multigenerational histories of population and land use change and integrate them with spatially explicit information on land cover, soil quality, climate, capital formation, market prices and participation in government programs. This synthetic approach will allow for the testing of the hypotheses from both household-centered and vulnerability-assessment theory about the impact of population loss on livelihood strategies and adaptive capacity, and contribute to the broader science of identifying thresholds of change in human-environment systems.